Can Civil Society Organizations Solve the Crisis of Partisan Representation in Latin America?
Kathryn Hochstetler Elisabeth Jay Friedman
ABSTRACT
This article takes up the question of whether civil society organiza- tions (CSOs) can and do act as mechanisms of representation in times of party crisis. It looks at recent representation practices in Argentina, Bolivia, and Brazil, three countries where political par- ties have experienced sharp crises after several decades of mixed reviews for their party systems. At such moments, any replacement of parties by CSOs should be especially apparent. This study con- cludes that the degree of crisis determines the extent that CSOs’ representative functions replace partisan representation, at least in the short term. Where systems show signs of re-equilibration, CSOs offer alternative mechanisms through which citizens can influence political outcomes without seeking to replace parties. Where crisis is profound, CSOs claim some of the basic party functions but do not necessarily solve the problems of partisan representation.
he words political party and crisis have become regular compan- T ions in the study of Latin American politics. The empirical referents range from longer-term patterns, such as the instability of partisan iden- tification, to short-term crises resulting in the collapse of party systems. Political parties’ abilities to meet performance, legitimacy, and repre- sentation expectations have all been questioned (e.g., Hagopian 1998; Mainwaring et al. 2006; Pearce 2004). In other contexts, scholars have argued that such crises should lead—and have led—to a transformation of mechanisms of representation: “citizens across the world have shifted from older and traditional forms of representation, such as polit- ical parties and unions, to ‘newer’ modes, such as social movements, informal citizen groups and nongovernmental organizations” (Chandoke 2005, 308). This transformation implies a radical move beyond the parameters of the state-society relationship suggested by pluralism, wherein competing interest groups lobby politicians but parties retain a monopoly on repre- sentation (Dahl 1961). We call such arguments the “crisis of representation hypothesis” because the change in modes of representation is explained primarily through the failure of traditional representation systems (Bartolini
© 2008 University of Miami 2 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 and Mair 2001; Chandoke 2005; Lawson and Merkl 1988; Lawson and Poguntke 2004). The implication is that the scope of nonparty representa- tion is an inverse function of the quality of partisan representation. The crisis of representation hypothesis was first developed in the European party literature, and clearly does not explain developments in all of Latin America. In some countries, party system meltdown has led to the reinvention of populist mediation by such leaders as Alberto Fuji- mori in Peru and Hugo Chávez in Venezuela. Citizens in such countries do not have the option of choosing between partisan and “newer” modes of representation, which we refer to collectively as civil society organizations (CSOs). CSOs are weak or have been largely crowded out. In other Latin American countries, however, party crisis does coexist with significant civil society self-organization. The crisis of representa- tion hypothesis prompts us to examine developments in those countries more closely, to see whether CSOs are able to play more extensive rep- resentational roles when parties are weakened. To do so, we compare Argentina, Bolivia, and Brazil, three coun- tries where political parties have experienced sharp crises since 2000 coupled with an upsurge of CSO activity. If the crisis of representation hypothesis has any relevance in Latin America, it should be evident in such countries. Our cases represent a spectrum of crisis intensity, rang- ing from total party system collapse (Bolivia) to partial collapse (Argentina) to delegitimation of the governing parties (Brazil). The hypothesis suggests that there should be more evidence of a substitu- tive shift to the newer CSO forms of representation in Bolivia, and pos- sibly Argentina, than in Brazil. While the hypothesis assumes that CSOs can actually act as repre- sentational mechanisms, this is far from an accepted understanding of their possible roles. Therefore, we begin by offering a conceptual explo- ration of CSOs as modes of representation, using the traditional mecha- nism of political parties as a point of comparison. The discussion also provides the foundation for assessing the different levels and kinds of representation that CSOs might perform. Insider strategies of CSO rep- resentation largely serve as complements to partisan representation, while outsider strategies seek to replace it. Thus the crisis of represen- tation hypothesis anticipates that CSOs will move toward outsider strate- gies as parties founder. The empirical sections of this article characterize the nature and extent of the recent party crises in these three countries, as well as the roles of CSOs during and (if applicable) after the crises. We conclude that there is general support for the crisis of representation hypothesis that the depth of party crisis is related to the scope of nonpartisan rep- resentation. Only in cases of severe crisis is the replacement thesis par- tially upheld; in Bolivia, CSOs really are using outsider strategies to HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 3 claim direct party roles alongside parties, including agenda setting, effective decisionmaking, fielding candidates, and checking or revoking executive power. In the more circumscribed crises in Argentina and Brazil, CSOs do not replace parties as mechanisms of representation— and they do not even try to, except for a brief period in Argentina at the height of the crisis in 2001–2, when the nation cycled through five pres- idents in just a few weeks. In both countries, however, even insider CSOs can carry out important representational functions.
POLITICAL PARTIES AND CSOS AS MECHANISMS OF REPRESENTATION
A curious disconnect exists between theoretical studies of representa- tion and most empirical studies. Theoretically, representation is a con- tested concept, with multiple meanings (e.g., Manin 1997; Pitkin 1967; Saward 2001). At its core, the concept demands that the values and interests of citizens in some way guide those who stand for them in col- lective political processes—functions that could be performed in many ways. Yet empirically, scholars have tended to restrict their focus to the world of parties, elections, and electoral systems, often without com- ment (e.g., Kitschelt 2000; Powell 2004; Przeworski et al. 1999). This article challenges that empirical restriction. It argues that CSOs also can perform what Bartolini and Mair call the political integration functions of partisan representation: articulating and aggregating interests and integrating and mobilizing citizens (Bartolini and Mair 2001). Thus, although CSOs do not generally engage in the specific democratic elec- toral form of representation particular to parties (Peruzzotti 2006; Rehfeld 2006), they fulfill a range of other functions (Friedman and Hochstetler 2002; Fung 2003; Gurza Lavalle et al. 2006). Understanding both parties and CSOs as mechanisms of representation requires careful attention to their structural similarities and differences.
Representation by Political Parties
Political parties are the primary channel for representation in democra- cies. Following V. O. Key’s influential formulation, parties perform three major groups of functions, each of which is central to representation: parties in the electorate, parties as organizations, and parties in govern- ment (Dalton and Wattenberg 2000, 5–10). Similarly, parties may fail at one or some subset of these functions. In their role of relating to the electorate, parties serve as interpreters for the voting public, presenting different ways to understand political issues and helping voters to determine their general political interests and identities. In doing so, some parties engage in extensive organizing, 4 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 reaching deep into civil society. But many serve exclusively as electoral vehicles that become dormant between electoral cycles. Party-society ties have long been weak in Latin America, but they continue to weaken in the posttransition democracies. Even where party systems were once based on more programmatic and mass-participatory linkages, those ties have now loosened. Personalist, patron-client, and charismatic linkages prevail. The result is party-society relations grounded largely in parties’ electoral ambitions, a thin relationship with little interest and legitimacy for citizens (Pearce 2004; Roberts 2002; Taylor-Robinson 2001). Similar processes are at work in Europe, albeit from higher historical levels of integration (Dalton and Wattenberg 2000; Lawson and Poguntke 2004). In both old and new democracies, citizens are increasingly consumers rather than subjects of politics (Bartolini and Mair 2001, 333). Parties as organizations perform the functions of recruiting elites for office and running their campaigns. Articulating interests and aggre- gating them into broad programs of government are related functions. Latin American parties have generally performed these party functions comparatively well, with the notable exception of the programmatic function. Many Latin American parties lack coherent programs of gov- ernance (Roberts 2002), which undermines a number of their other party roles. If candidates succeed, parties continue to mediate citizens’ interests, now in their third role as authoritative decisionmakers. They organize the legislative process and oversee policy implementation. Parties have maintained a monopoly over most governing roles, especially through rules that require candidates to be partisan. In democracies, electoral rules set procedures for periodic re-evaluation by voters and the possi- bility of replacement, establishing accountability. Much of the evaluation of parties as mechanisms of representation analyzes the consequences of variations of these rules (e.g., Carey 2003; Mainwaring and Scully 1995). The ability of Latin American parties to carry out these governing functions well has varied quite a bit over both time and space (Main- waring 1999, 2006).
Representation by CSOs
We begin with a deliberately broad definition of CSOs: they are volun- tary associations that promote the interests of citizens in a variety of ways. This study is primarily interested in their actions that promote the values and interests of citizens in collective decisionmaking processes. As agents of representation, CSOs differ from parties in two fundamen- tal ways: they are not chosen by an electorate, and they do not govern. Therefore, many analysts are hesitant to use the term representation as a description of CSOs’ activities (Mainwaring 2006; Peruzzotti 2006). HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 5
This study, by contrast, follows Rehfeld in arguing that both common sense and theoretical understandings of representation require a broader definition than that based on the democratic electoral process. Rehfeld sees representation wherever there is “an audience’s judgment that some individual, rather than some other, stands in for a group in order to perform a specific function” (Rehfeld 2006, 2). For example, NGOs seek to represent the interests of those who have “no say in the selection of their representatives” (such as the prisoners of war for whom the Red Cross advocates) and who may not even be human (such as the environment) (Rehfeld 2006, 1). Audience recognition is the key to identifying instances of representation, rather than institutions. In the case of CSOs, this study argues that both citizen and government audi- ences—and many CSOs themselves—often accept them as representa- tives of collective interests and values (Gurza Lavalle et al. 2006). For CSO representation to be democratic, there must be some recognition of them from the citizens the CSO claims to represent. The CSO behaviors recognized as representative often parallel the party functions just outlined, albeit not as closely linked to the electoral process. As CSOs perform these functions, they are less restricted (and less enabled) by formal rules and procedures than are parties. While most political systems have formal rules for party formation, CSOs are typically freer to self-constitute around particular interests. But they also may be subject to rules on their formation or action, particularly if it involves external funding, political lobbying, or policy formulation or implementation. The comparative flexibility of CSOs’ aggregation processes creates a more amorphous relationship of representation with their subjects (see Roberts 2002). In linking to the mass public—conceptualized as the electorate for political parties and citizens for CSOs—CSO and partisan representation take different shapes. Party-based representation aspires to link to a broad public, usually most of the adult population. This universal qual- ity of parties’ reach is cited as a reason that they cannot be replaced by CSOs, many of which target much more specific constituencies (see, e.g., Mainwaring 1999, 13). CSOs have a different scope advantage: their activities are more likely to be continuous over time, while parties are most active in electoral cycles. CSO campaigns and activities can offer citizens symbols of identification and loyalty, education, opportunities for political claimsmaking, and the simplification of choices at all times. During electoral campaigns, CSO efforts may intersect those of parties; helping voters identify the “environmental candidates,” for example. In relation to citizens, each mechanism has a specific basis on which it makes representational claims. Parties justify their claims to rep- resentation based on their membership numbers or electoral results. This representational claim is relatively straightforward: it translates a 6 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 tangible and quantitative measure of support into political power or, at least, political preferences. CSOs may struggle to justify their actions because they rarely “measure” their support through elections or, in many cases, membership. CSO support may be tracked by the numbers of people who support their campaigns; for example, by responding to their calls for street protests. Public opinion surveys sometimes report public approval of particular movement sectors, CSOs as a whole, or movement values. The voluntary quality of CSOs means that a rise in their numbers indicates people’s increasing reliance on them for politi- cal linkage, although the emergence of new CSOs may also indicate that citizens or members find the existing ones inadequate.1 As for CSOs’ organizational roles, there is little controversy about the claim that CSOs articulate political interests; it is the only claim acknowledged by Dalton and Wattenberg (2000, 8), for example. Most CSOs see this as their central function, claiming to represent crucial identities or ideas neglected by other mechanisms. These may be widely supported ideas, although CSOs often introduce new ideas into political systems, such as insisting on human rights protections in authoritarian regimes (Peruzzotti 2006). In Latin America, CSOs are also active in interest aggregation, elaborating large-scale programs for governing, understanding of fundamental rights, development strategies (especially alternatives to neoliberalism), and institutional reforms (see, e.g., Bonner 2005; Peruzzotti 2002; Van Cott 2005). CSOs have even carried out the functions of recruiting and training political elites, alone or in conjunction with political parties. Current presidents recruited to national stature through CSOs include Brazil’s Luis Inacio Lula da Silva, a unionist; and Bolivia’s Evo Morales, leader of the indigenous move- ment and the coca growers’ association. Thus, both parties and CSOs can carry out important representation in the political system, articulating and aggregating interests and inte- grating and mobilizing people. But only political parties have a repre- sentational role that allows them directly to make authoritative decisions for society as a whole beyond their specific constituencies: they govern, in government. CSOs promote the interests of their respective subjects in the public sphere instead—that is, the sphere of collective discussion and debate outside of formal political decisionmaking (although they may, of course, be present to lobby within state institutions) (Fung 2003, 524–26). This public sphere has historically been underdeveloped in Latin America, but it has increasing relevance. It is a space where asso- ciations can put issues on the public agenda, introduce alternative polit- ical practices, and even develop new institutional formats for address- ing their issues (Avritzer 2002, 7; see also Fung 2003, 526–29). While they are not institutionalized forms of decisionmaking, CSOs’ actions in the public sphere can perform some of the same functions of HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 7 parties in government. In especially dramatic form, mass protests partly organized by CSOs are the new poder moderador for South American presidents, sanctioning their ability to continue governing or crippling them before their “fixed” terms end (Hochstetler 2006, 403). CSOs also can exercise societal accountability functions, monitoring the respon- siveness of government actors and the quality of government adminis- tration (Smulovitz and Peruzzotti 2000). Returning to the crisis of representation hypothesis, we propose two broad possible characterizations of the representation carried out by CSOs. These are grounded in a distinction between “insider” and “out- sider” CSO strategies (Smith and Korzeniewicz 2007). As insiders, CSOs use institutionalized strategies for influencing politics that leave partisan representatives in control of political decisions. This characterization remains within the boundaries of the state-society relationships expected by pluralist and related theories, where interest groups, lob- bies, and nongovernmental organizations add an additional layer of mediation between citizens and their elected representatives without displacing the latter. Insider CSO-based strategies support or comple- ment partisan representation; citizens use them because of their differ- ent scope advantages or because nonparty participation may be a better choice for expressing certain kinds of preferences outside the electoral season (Fuchs and Klingemann 1995; Kitschelt 2003). The crisis of representation hypothesis anticipates a more dramatic break with partisan representation when the quality and dependability of partisan representation are very low. In other words, the larger polit- ical context sets the strategy, rather than, say, the issue focus of the par- ticular CSO. Under these conditions, the hypothesis will expect CSOs to use outsider strategies that directly challenge or replace institutional decisionmaking. In the streets or in other public spaces, CSOs will use their mass power to overturn political decisions they do not support or to insist on particular outcomes. Elected representatives may still under- take the votes that make such outcomes the formal rule of the land, but they will do so under duress, and even against their own interests and preferences. CSOs in this outsider role may seek to institutionalize their power to represent interests in nonelectoral ways, or may seek mecha- nisms of direct democracy that further undercut the authority of elected representatives. Which of these characterizations best fits CSO activities in Latin America? To answer, we examine recent evidence about how well both parties and CSOs have performed their mass public, organizational, and governing functions in Argentina, Bolivia, and Brazil. The party litera- ture provides clear lists of indicators that we will use to assess the level of party crisis. These include survey data on disaffection with parties in general, erosion of partisan identification, electoral turnout figures, 8 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 increased electoral uncertainty, and increased voting for nontraditional parties and candidates. For CSOs, we will examine both the emergence of new organizations—especially new kinds of organizations—and the fortunes of existing organizations. We will present evidence of changing levels of support among different potential audiences for particular cam- paigns and organizations, as well as any available survey data. Last, we will track CSOs’ efforts to articulate and aggregate the interests of broad constituencies. Throughout, we will be seeking to categorize their strategies as insider or outsider in orientation. We move through the cases in order from the greatest to the least amount of party crisis, expecting CSOs to display successively more insider-oriented strategies.
PARTY CRISIS AND EMERGING REPRESENTATION IN BOLIVIA
Crisis and change in Bolivian politics are not new. Very high levels of institutional volatility culminated in South America’s only revolution in 1952. The MNR party led that revolution and its overturning 33 years later, putting the country on an unprecedented path of market eco- nomics and pacted liberal democracy. The MNR competed with the ADN and MIR parties in hard-fought elections; but then, shifting coali- tions of pairs of them joined in a governing pact that preserved the same basic policies for two decades (Mayorga 2005). These parties, and their steep decline after 2000, define Bolivia’s party crisis, the most severe of the three cases, albeit still within the limits of constitutional democracy.
Party Representation and Crisis
Electoral volatility and the rise of new parties are two of the basic com- ponents of party crisis, and both are fully evident in Bolivia. The MNR, ADN, and MIR won 73.4 percent of the popular vote in 1985, as Boli- vian politics was reconstructed after military rule and rampant hyperin- flation. From that peak, their share of the national vote began to decline, gradually and then more quickly. In 2002, their votes slipped below 50 percent, to 42.18—but Bolivia’s allocation rules still left them with 52.85 percent of the seats in the Chamber of Deputies (Mayorga 2005, 156), allowing them once again to select a president from their number.2 In 2005, only the MNR among the three ran candidates under its own name, winning just 6 percent of the vote. Other traditional politi- cians joined in a new political force, Podemos, which earned 29 percent of the vote.3 All, however, were swamped in the tidal wave of the MAS party, which won unprecedented majorities in the Chamber of Deputies (54 percent of the popular vote for the proportionally allocated seats and 64 percent of the uninominal seats) and won the presidency with HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 9
Evo Morales in the first round, also unprecedented (Corte Nacional Elec- toral 2005). Seven months later, in elections for a new constituent assembly, Podemos’s vote was cut nearly in half, to 15.3 percent, although it gained 23.5 percent of the seats and constituted the largest opposition force. MAS stayed just above 50 percent, with 50.7 percent of the vote and 53.7 percent of the seats (Ramiro Ballivián 2006, 132). Even before 2005, Bolivia was second in the Andean region in elec- toral volatility and third in vote share to new parties (Mainwaring et al. 2006, 19); it can only have moved “up” in the rankings. Indeed, by Mainwaring et al.’s definition of party system collapse—which is that “new parties gain more than 45 percent of the votes over the course of two consecutive lower chamber elections” (Mainwaring et al. 2006, 21)—the posttransition Bolivian party system has collapsed. This col- lapse rests on a similar drop in citizens’ regard for their parties: from 20.4 percent expressing “some or a lot of respect” in 1997 to just 6 per- cent in 2003 (Latinobarometer figures, cited in Mainwaring et al. 2006, 17). The crisis is also apparent in that the two presidents before Morales were forced to resign from office early, in the wake of large protests. MAS and Morales are emphatically not repackaged politicians from the traditional parties, although they are not wholly outsiders either. Morales and others from MAS were elected to Congress in 1998, so they now have a decade of institutional experience. From the beginning, their electoral roles were closely linked to nonelectoral political roles, however. MAS is officially the “political instrument” of a set of mostly rural and indigenous unions and movements. Morales has headed the coca growers’ union, CSCB, since the 1980s, and was re-elected its pres- ident in February 2006 after being elected president of the country (Gamarra 2007, 12). Morales was expelled from Congress in 2002 after leading violent protests against coca eradication (Barr 2005, 69, 72), although the incident seems to have ultimately helped him tap into the broader opposition to politics as usual. Thus Morales and MAS straddle the line between electoral and nonelectoral forms of representation, and the success of their first administration will depend on how well they can ride the two horses simultaneously.
Emerging Forms of Representation: Direct Democracy
The La Paz newspaper La Razón prints a daily Marchodromo telling who closed down the streets the previous day and why, and who is scheduled to march on the current day. This daily feature reflects the routinization of street protest and road blockages as political strategies in Bolivia. As an earlier generation of union-led contention declined with the peak labor organization COB in the 1980s, indigenous move- 10 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 ments gradually moved into the vacuum with their road-closing marches (Yashar 2006), now copied by their political allies and foes alike. Protest marches are often seen as inputs into representational poli- tics. In Bolivia, however, protest marches have proven to be strong enough to force political outcomes clearly opposed by elected repre- sentatives in their institutional roles. Days of violent protests by neigh- borhood organizations and civic committees in Cochabamba in 2000 made the government undo privatization of the city’s water supply system (Domingo 2005). A few years later, massive protests against nat- ural gas exports forced president Gonzalo Sánchez de Lozada out of office early; pushed his successor, Carlos Mesa, to hold a national ref- erendum on hydrocarbons; and then forced Mesa’s resignation. Mesa was, in reality, squeezed between an indigenous-led coalition and an emerging countermobilization for regional autonomy from the lowlands. Before the dust had settled on Mesa’s resignation, protesters also rejected two individuals in the constitutional line of succession before accepting the next one—along with promises of early general elections, an elected constituent assembly, and direct elections of departmental leaders. The Congress even summoned the supermajorities to make nec- essary constitutional changes, drawing on votes from representatives whose parties were predictably decimated in those subsequent elec- tions. Protest clearly became a strategy used by CSOs across the politi- cal spectrum. Bolivians from all parts of society freely acknowledge that their pol- itics has come to be fairly directly set by large social movements—rep- resentation outside of the electoral system—rather than by institutional actors (see Barr 2005; Domingo 2005). In the introduction to its pro- posed constitutional draft, conservative Podemos explained that the constituent assembly its members had long resisted was necessary “because the Bolivian State finds itself out of harmony with civil society. The political system does not adequately channel the representation, much less the effective participation, of broad sectors of the population, of the indigenous peoples, of women, or of the common citizen” (Podemos n.d., 3). Vice President Alvaro García Linera’s sociological training was evident when he analyzed the existence of two levels of “democratic institutionality” in Bolivia: the first, the level of representa- tive institutions, and the second of “the participatory force of the social collectivities: unions, professional organizations, social movements, civic committees” (REPAC 2007a, 79). Beyond the indigenous movements and unions that helped sustain his party, MAS, García Linera drew attention to the unprecedented mobilizing capacity of the “new right” in the regional civic committees. These committees collected half a million signatures in support of a ref- erendum for regional autonomy in 2005–6, and could put tens of thou- HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 11 sands of people into the streets to pressure the government. In the state of Santa Cruz and other parts of the so-called crescent (medialuna) of wealthier lowland states, the committees have recently gone so far as to declare independence from the central state (not yet successfully). Such claims are probably the most significant outsider strategy possi- ble, a wholesale rejection of a system of political representation. This extension of the direct protest strategy to the right-wing as well as the leftist CSOs signifies the deeper hold of noninstitutional representation strategies in Bolivian politics compared to those of Argentina and Brazil. It is interesting that García Linera and Podemos concurred in hoping that constitutional changes could bring demands into institu- tionalized channels. The perennial demand of indigenous CSOs for constitutional changes that would “refound the Bolivian Republic” shows that they, too, wanted to institutionalize their access to political power, and not necessarily in ways that strengthened electoral representation. A 1990 march by lowland indigenous populations for “land and dignity” marked a relatively modest beginning for what became, a decade later, a demand for wholesale constitutional change. Renewed protests in 2000 and 2001 earned an Act of Understanding and creation of a com- mission of notables (consejo ciudanano), which wrote a proposal for constitutional change that largely ignored the suggestions of rural unions and indigenous organizations. In response, highland and low- land indigenous populations joined in a collective protest march in May 2002, from Santa Cruz to La Paz. This time, having concluded that tra- ditional politicians would not make the changes they wanted, they asked to write the new constitution themselves (Betancur 2004, 39). The tumultuous politics of the next years—when the protests pushed out two presidents and rewrote Bolivian posttransition political history—reflected complex struggles over exactly what institutionaliza- tion would mean. As Mesa tried to complete his term in the presidency, his administration shepherded a set of constitutional changes through Congress that allowed citizen groupings (agrupaciones ciudadanas) and indigenous peoples to gain registration with the National Electoral Tribunal; this would allow them to run candidates against party candi- dates. Another change that allowed citizen legislative initiatives was used by the Santa Cruz civic committee to put on the ballot a referen- dum on regional autonomy, along with the referendum about whether to hold a constituent assembly. While these changes did break the monopoly of the political parties, indigenous groups and other CSOs rejected them as insufficient, saying that they retained a party logic that did not fit with social movements (Betancur 2004; Ojeda 2007; Pacto de Unidad 2004). Indeed, the most prominent of the “citizen groupings” was Podemos, which, as a repackaged ADN, supports that argument. 12 LATIN AMERICAN POLITICS AND SOCIETY 50: 2
Instead, indigenous groups supported representation options that would guarantee that constituents be an ethnic mirror of society. An early proposal, debated with Morales’s coca growers’ union and MAS deputies as participants, called for a triple-list system: one indigenous, one women, and one “other cultures”—none party-based (CSUTCB et al. 2004, 6). A few months later, a wider indigenous coalition formed the unprecedented Pacto de Unidad (Unity Pact) of highland and lowland indigenous groups. They presented Congress with a plan for convoking a constituent assembly that would join selection of participants through universal votes for candidates from party, citizen group, and indigenous peoples’ lists (221 of 248) with 10 lowland indigenous representatives, 16 highland indigenous representatives, and one Afrobolivian represen- tative. The latter 27 participants were to be selected by community “practices and customs” (usos y costumbres, language from the 1994 constitutional changes) rather than by elections (Pacto de Unidad 2004, 19). MAS offered some support for this proposal, but sacrificed it for universal electoral representation in its negotiations with Podemos to achieve the two-thirds supermajority needed to call a constituent assem- bly. MAS’s action split the Unity Pact, with rural unions but not indige- nous groups accepting the trade-off (Carrasco 2007). The constituent assembly process itself has encouraged direct par- ticipation. Its committees traveled across the country holding depart- mental consultations. In addition, committees were to hold weekly public audiences to hear citizen proposals, and the Office of the Presi- dency gathered and published 80 proposals for constitutional reform (REPAC 2007b). Twenty-five came from party and other public organi- zations, 29 from various social movements, 9 from NGOs, 4 from the regional civic communities, and 3 from the business community. Virtu- ally all of the Unity Pact reunited to put in a joint proposal that repeated their earlier demand that “the indigenous and original nations and peo- ples and campesinos should exercise direct representation in the differ- ent levels of government on the basis of their practices and customs and on the principle of communitarian democracy” (Asamblea Nacional 2006, 4). They also asked to institutionalize the revocatory power against presidents they had been exercising from the streets. Some of the pact’s member organizations have made autonomy for indigenous communities even more central to their individual proposals; the indige- nous organization CONAMAQ has called for a return to the traditional Aymara ayllu, or community administration (CONAMAQ 2005). Electoral processes have brought Bolivians a president who uses first-person plural verb forms to talk about indigenous people. MAS has moved already to pass legislation on key indigenous and social move- ment demands for nationalization of energy and for land redistribution, drawing on more large protest marches to buttress its thin legislative HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 13 majority in the lower house and minority position in the upper one. MAS deliberately waited to formulate its constitutional proposals until May 2007, when the societal proposal stage was complete, and sup- ported many of the indigenous and social movements’ proposals, including the one on revocatory power. Even so, it did not include the call for direct indigenous representation (REPAC 2007b, 39–40). MAS talks frequently of the need for indigenous and other forms of auton- omy; and autonomy—indigenous, departmental, regional, and other- wise—is one of the major themes of this constituent assembly. Whether MAS means in its use of the term what CONAMAQ and others do is far from clear. It is equally unclear if the net balance of substantive representation under MAS and other new political forces will resolve the larger crisis of procedural electoral representation. When MAS delivers for its con- stituencies, those of Podemos respond with their own system-challeng- ing demands. CSOs themselves are divided in many important ways: civic committees from indigenous, rural unions from indigenous, low- lands from highlands. Perhaps the worst possible outcome already has briefly appeared: in June 2007, in the heat of negotiations over a new constitution, MAS and Podemos debated proposals in institutional set- tings to the point of incivility—shouting, breaking glass, shoving—while their street supporters have just narrowly avoided even more violent confrontations. At best, the Marchodromo is likely to continue to be a daily feature.
PARTY CRISIS AND ALTERNATIVE REPRESENTATION IN ARGENTINA
Argentine society erupted in full-fledged rebellion against its political class in mid-December 2001. Under the banner “Throw them all out” (Que se vayan todos), groups ranging from human rights champions Mothers of the Plaza de Mayo to movements of the unemployed took to the streets to protest the failure of the formal mechanisms of repre- sentation. By 2002, it looked as though a diverse set of nonpartisan CSOs were constructing an alternative, outsider arena for horizontal and localized political participation. But how profound has this change been? The Argentine party system is in flux, but not all parties are in crisis. And although civil society has offered alternative forms of repre- sentation, it is not replacing party politics.
Party Representation and Crisis
Historically, the Argentine party system wavered between two-party and moderate multiparty. Besides the dominant parties, the Peronist Party 14 LATIN AMERICAN POLITICS AND SOCIETY 50: 2
(PJ) and the Radical Civic Union (UCR), significant party representation existed on the left and the right. The parties’ collective inability to resolve the economic catastrophe that by mid-2002 had resulted in 50 percent poverty and 25 percent unemployment, however, sent the polit- ical system into a tailspin. Between 1997 and 2002, Argentines’ overall confidence in political parties dropped precipitously: 86.2 percent, the largest decline in Latin America, ending with a mere 4 percent of the electorate evincing confidence in parties (Latinobarometer data, cited in Sánchez 2005, 457). This distrust was reflected in the 2001 elections, when only 30 percent of the electorate chose to vote for the two major parties, and 15 percent nullified their ballots or left them blank (Calvo and Escolar 2005, 211–12). While these numbers suggest that parties entered severe crisis, a dif- ferent story emerges from detailed analyses of electoral behavior. Most of the crisis was in the UCR. The decline of national electoral power for the combined two major parties in the first decade of democracy (from 92 to 67 percent of the presidential and 86 to 65 percent of the legisla- tive vote) was mainly due to UCR’s performance (Torre 2003, 649). Although it still controls about one-fourth of the governorships and municipalities, the UCR “has effectively disappeared as an important national-level political formation representing the middle class—an iden- tity and role dating as far back as the 1890s” (Sánchez 2005, 472, 467). Beyond the UCR, there was some quick recovery in electoral par- ticipation from the low point in 2001. In the presidential elections of 2003, the abstention rate was 22 percent lower than in 2001, while blank and null votes, at just 2.6 percent, were at their lowest since 1983 (Sánchez 2005, 468). Although a host of presidential candidates divided the vote, half were Peronist, and the other half had their origins in the UCR (Sánchez 2005, 461–65, 471). The PJ—which came in first and second in the 2003 presidential elections and won handily in 2007—has risen phoenixlike from the ashes. Torre (2003, 659) argues that there is a “hard core” of PJ supporters, made up of about 37 percent of the electorate, who will vote for the party regardless of the policies it supports in office.4 Historical loyalty to the party of Juan and Evita Perón is reinforced by material rewards. The party’s successful transition from union-based to clientelistic links during the neoliberal 1990s has meant that even poor voters continue to have limited access to resources (Auyero 2007; Delamata 2004; Levitsky 2001). Institutional design also contributes to its staying power: the PJ gains a disproportional number of seats, particularly in the Senate, because of its overrepresentation in small provinces (Torre 2003). The current party system as a whole remains “in a highly fluid state” (Sánchez 2005, 473). Torre (2003) argues that voters from the center- right and center-left have been particularly volatile, and in ways that HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 15 simultaneously enhance and damage party-based democracy. More demanding citizens, educated by the human and civil rights movements of the 1980s and 1990s, refuse to be loyal to parties that are corrupt, inefficient, and ideologically incoherent in the face of crisis. Still, the party system and political parties are not in complete crisis: the PJ sur- vived the general upheaval of 2001–2. Its ability to channel material benefits to its supporters also has been an incentive for CSOs to con- sider insider strategies in regard to the government.
Emerging Forms of Representation: Autonomy, Cooptation, and Cooperation
Civil society was primed for revolt in 2001. In the 1990s, the economic dislocations caused by Menem’s neoliberal reforms wreaked havoc in the provinces. The response established a pattern that would mark the decade. Seemingly entire cities rose up in Salta, Jujuy, and Neuquén provinces, attacking government offices and cutting off major trans- portation arteries in what were characterized as spontaneous demon- strations against plummeting living and working conditions.5 In these areas and Greater Buenos Aires, movements of the unemployed coa- lesced, demanding that provincial and national governments provide welfare and employment (González Bombal 2003; Svampa 2005; Svampa and Pereyra 2005). At their most radical, these movements sought to break with forms of representation that they found to be fundamentally based in depend- ent, clientelistic relations. They promoted horizontal relations and deci- sionmaking through direct democracy (Sitrin 2005, iv–v, 27). The rejec- tion of electoral and party-based representation in favor of autonomous political processes became widespread as the crisis broadened and deepened. The new movements were inspired by their predecessors. The gen- eration of human rights groups that opposed the authoritarian “Dirty War” (1976–83) continued its work after the transition to democracy. That movement inspired a second (if overlapping) generation of rights proponents, who fought against impunity while strengthening demo- cratic institutions to protect human and civil rights (Peruzzotti 2002). Even closer to home for the new social movements’ rejection of “poli- tics as usual” was a new union movement, led by the antineoliberal Argentine Workers’ Congress (CTA) (Bielsa et al. 2002, 71). Reflecting shifts in social and political organization, the CTA offered a new form of representation for a membership that went beyond traditional defini- tions of workers. It refused the partisan identification of the old PJ- linked labor confederation and held direct elections of its leadership. It welcomed nonunion members, seeking to be an “organizing tool that 16 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 would unify the whole world of work, including informal and unem- ployed workers” (Bielsa et al. 2002, 71). Furthermore, it sought to link workplace-based and new forms of organizing, declaring as its target for mobilization “the new factory which is the neighborhood” (Camarero et al. 1998). At the peak of political crisis, many of these actors came together in their critique of the legitimacy of parties and the political class. This was most strongly articulated in the societywide demand to “throw them all out.” This demand was first asserted at the massive demonstrations against Fernando de la Rúa’s government in December 2001 (Lozano et al. 2004). It was taken up in public spaces, from cafés to airports, by those who loudly chastised political figures, holding them accountable for the failure of the political system. In the wake of the protests and the resignation of both de la Rúa and the following interim president, autonomous organization exploded. A largely middle-class phenomenon, the neighborhood assemblies fol- lowed in the steps of the unemployed movements (Bielsa et al. 2002, 24), engaging in protest action—including the famous pot banging (cacero- lazos), roadblocks (piquetes), and “outings” (escraches) of public fig- ures—as well as horizontal decisionmaking on local issues. The assertion of direct democracy in lieu of representative democracy (Bielsa et al. 2002, 15, 31) was concretized in the hours spent in open discussion. Some workers occupied and restarted shuttered factories, expanding into a movement of “recuperated” factories (Korol 2004; Lavaca 2004). Social movements offered opportunities based on gender as well as class; through them, both poor and middle-class women found the “possibility of reaching public space” (Sanchís 2006; Di Marco 2006). As they grew, these movements combined forces. Their shared out- sider strategies (self-organization, horizontal deliberation) against common enemies (the political establishment, neoliberal economic policies) articulated the demands of a broad constituency. Or, in the words of a protest chant, “road blockade and pot banging, the struggle is the same” (Piquete y cacerola, la lucha es una sola) (Bielsa et al. 2002, 101). There were differences in terms of their reception by their target audience, however: by 2002, the assemblies had twice the level of acceptance (80 percent) by the general population as the unemployed movements (40 percent) (Bielsa et al. 2002, 26). Institutionalized CSOs also found in the crisis an opportunity to advance their reform agenda, albeit through more insider approaches. Between 2000 and 2004, 16 NGOs were formed that focused on state reform (CIPPEC 2006). As a primary example, a group of professional civil rights NGOs decided to address the problems with the discredited Supreme Court. The group offered critiques of the court’s design and proposed reforms in the document “A Court for Democracy” (Aso- HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 17 ciación por los Derechos Civiles et al. 2002). In 2003, President Néstor Kirchner signed a decree accepting most of their ideas (Asociación por los Derechos Civiles 2005, I.1). Thus, a coalition of NGOs was recog- nized as having aggregated, and as now representing, societal interests by the audience they addressed, state decisionmakers. Those activists who focused on institutional reform did not claim to be replacing the democratic electoral representation of political parties. As Roberto Saba, executive director of the Association of Civil Rights (ADC), explained, “Human rights NGOs have no need to prove they are representing anyone. They are advocating for people’s rights.” He offered an illustration of their unique role, telling a story of testifying before Congress during a debate on a freedom of information bill. When a powerful senator sought to deny his legitimacy as an unelected inter- locutor, he replied, “I don’t represent any voters. I’m here to make sure the laws are complied with, to fulfill the constitution.” He argued that the groups’ work promoting individual rights will often be “counterma- joritarian” (Saba 2006). Even so, their work clearly promotes citizens’ interests that are crucial to democracies. Academic observers enthusiastically reported how the new actors used outsider strategies to challenge and reject partisan and institutional forms of representation (Svampa and Pereyra 2003). “The autonomous Argentine social movements . . . have begun to articulate a new and rev- olutionary politics . . . a politics of daily life” (Sitrin 2005, iii). The neigh- borhood assemblies offered an “irreversible” “new radicality . . . which has taken hold in the new generation” (Bielsa et al. 2002, 55).6 Contemporary developments, however, reveal a soberer picture of the current politics of representation. Facing rivals for the presidency in his own party, President Kirchner (2003–7) sought nontraditional allies, including human rights groups, progressive intellectuals, and unem- ployed movements. His actions to reverse the decades-old impunity for those responsible for the Dirty War, including a military purge, com- mended him to several human rights organizations. It seemed that recent issues would also be on the table: he accepted the court reform and refused to crack down on assemblies and unemployed workers (Gaudin 2005). Some groups cast their lot with a government they saw as willing to address their needs. In a striking symbol, Hebe de Bonifini, leader of the Mothers of the Plaza de Mayo, declared the annual 24-hour Resis- tance March unnecessary because “the enemy is no longer in the Gov- ernment House” (Boschi 2006). Two prominent leaders of neighbor- hood and unemployed workers’ movements left the executive committee of the CTA to enter the Kirchner government (Etchemendy and Collier 2007, 369). Such CSOs, some of which had continued to work closely with the Peronist Party during the height of the crisis 18 LATIN AMERICAN POLITICS AND SOCIETY 50: 2
(Auyero 2007), accepted one party as offering legitimate representation and worked toward strengthening its position. There was a material incentive for this support: Kirchner continued and augmented social assistance benefits. In December 2001, one percent of the labor force had government subsidies; by April 2003, 18 percent did (Sánchez 2005, 468-69). A 2005 estimate of national welfare plan beneficiaries was 1.7 million (Svampa and Pereyra 2005). Other CSOs, by contrast, offered a range of critiques of the Kirch- ner government, from outright rejection to more cautious evaluation. Some human rights activists made a distinction between Kirchner’s sup- port for commemorating the victims of the Dirty War, such as declaring the most notorious torture center a museum to memory, and the needs that remained for both truth and justice (Antokoletz 2006). Contempo- rary issues were eclipsed by the focus on the past as impunity for cur- rent human rights violators continued (Valente 2006; Walger 2006). Those who rejected the governing party and its charismatic president made their critiques through both partisan and nonpartisan vehicles. Sectors of the unemployed movement cooperated with or were led by far-left parties (Delamata 2004), as were several of the ever-dwindling neighborhood assemblies. The politically independent Coordinator Against Police and Institution Repression (CORREPI) used the court system, along with grassroots education, to fight against ongoing phys- ical repression by state security forces (Verdú 2006). The 2001 demand for a renovation of political life failed: no replacement of the political class occurred. The upsurge of innovative citizen participation was fleeting: “the axiom existed that ‘they should all go’ and really not one of them left. And the citizens rapidly returned to the most daily problems of the crisis, of their own social, economic, etc. problems” (Vezzulla 2006). The CTA reported in 2004 that nearly 90 per- cent of contemporary social protests attracted fewer than five thousand participants (Lozano et al. 2004). Some observers find clientelism responsible for this reflux of ener- gies. As the president of a housing movement argued, “hyperwelfariza- tion” steered many groups to “practices that generate dependency and a loss of dignity” (Jeifetz 2006). Kirchner relied on a range of strategies to “integrate, coopt, discipline and/or isolate” the unemployed move- ments (Svampa and Pereyra 2005, 107). Continuing disillusionment with parties was attested to by interviews in 2006, in which questions explor- ing CSOs’ representational roles were often answered by a strong dis- sociation from partisanlike claims to represent numbers of individuals. Although partisan leadership—especially from the everpresent PJ— remains in place, civil society continues to provide for some aspects of representation, if now in less dramatic form. On the institutional side, Saba (2006) noted an awareness of the need for civil society to profes- HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 19 sionalize articulation efforts, taking its insider role seriously; some formal sector unions have developed successful neocorporatist strate- gies (Etchemendy and Collier 2007). On the movement side, neighbor- hood, unemployed, and other grassroots groups continue to join together in protest, as perhaps most dramatically seen in the growth of the cross-class mobilization for the legalization of abortion, attributed by one participant partly to the encounters between unemployed women and feminists during the crisis (Schvartzman 2007). But even here, strategies are a mixed bag of insider and outsider approaches, as repro- ductive rights proponents struggle for legal as well as social change.
PARTY CRISIS AND EMERGING REPRESENTATION IN BRAZIL
Brazilian political parties have long been criticized as among the weak- est in Latin America, although the Workers’ Party (PT) stands out among them as unusually programmatic and disciplined (Hunter 2007; Samuels 2006). When extensive corruption charges against the PT emerged in 2005, they created a sense of crisis for the PT’s first national adminis- tration and the party itself. Because of the PT’s ostensibly unusual char- acter, its crisis also caused generalized questioning about party-based representation, although this never reached the extent of the debate in Argentina or Bolivia.
Party Representation, Corruption, and Crisis
In 1998, near the end of the tenure of Brazil’s first full-term civilian pres- ident, Brazilians stood squarely at the midpoint of Latin American coun- tries in their levels of confidence in their institutions as a whole: 20 per- cent of them expressed a great deal or quite a bit of confidence in their parties, between Uruguay’s high of 34 percent and the 15 percent of Ecuadorians and Venezuelans, at the bottom of expressed confidence levels (Latinobarometer numbers, cited in Turner and Carballo 2005, 181). By 2005, only 13 percent of Brazilians expressed similar levels of confidence, but they were still very near the Latin American average of 13.8 (Latinobarometer numbers, cited in Sánchez 2005, 457). Measures of partisanship—as expressed in individuals’ self-identifi- cation with a particular party—tell a similar story. Most Brazilians do not consider themselves partisans of any party, but the vast majority who are partisans have traditionally been committed to the PT. Preference for the PT “grew more or less consistently from 6 percent in 1989 to 24 percent in late 2004” (Samuels 2006, 5). In August 2005, however, as the corrup- tion scandal peaked, self-declared PT preference dropped to 18.5 per- cent, and still stood at 18.8 percent in September 2006, even as the PT 20 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 president, Lula, won re-election (DataFolha 2005, 2006). Other parties did not pick up these former PT partisans; the PMDB’s hold, for exam- ple, has gradually declined to just 7 percent of the committed partisans, and other parties hover at 5 percent and below (Samuels 2006, 5). In several ways, then, survey data support the argument that 2005 brought political parties, and especially the PT, to a crisis in their rela- tions with the electorate. Voters do not identify with the parties or hold them in high regard: 57 percent of Brazilians interviewed in August 2006 could not even remember the candidates for whom they had voted in the Chamber of Deputies and the state-level legislative assembly elec- tions in 2002 (DataFolha 2006). In 2005, only 42 percent of Brazilians thought that it was impossible to have a democracy without political parties (Latinobarómetro 2005). In the October 2006 elections, the first requiring parties to receive at least 5 percent of the valid votes for fed- eral deputies in order to receive full party status, there was a 46 percent turnover rate in the lower house. Yet the same four parties that have dominated Brazilian politics since the early 1990s continue to draw the most votes, showing little of the electoral volatility or uncertainty con- sidered to be an indicator of antiparty sentiment. PSOL, a new party that broke off from the PT in 2003 and was the most likely bearer of a protest vote, did not meet the 5 percent threshold. August 2005 was clearly the peak of the party crisis in recent Brazil- ian politics. This was especially evident for the PT, which had achieved the presidency and unusual levels of partisan support, only to see the latter slip away very quickly. Yet the extent of party crisis was limited in comparison to Argentina and Bolivia. About ten thousand people demonstrated to demand Lula’s impeachment that August, but their numbers did not approach the millions who helped remove President Fernando Collor de Mello in 1992. They were met, moreover, by an equal number of counterprotesters supporting Lula. This sharpest moment of crisis was shaped by ongoing party logics. Lula’s disap- pointed historical supporters in CSOs did not mobilize against him, partly because it would have meant allying with the PSDB party against the PT (Barbosa 2006). In addition, the people who gathered to demand Lula’s impeachment were organized by the breakaway PSOL and its par- tisan supporters. Thus, even at the peak of the crisis, Brazilian CSOs did not try to organize direct challenges to the parties’ control over govern- ing functions, except for some members’ choice to switch partisan sides.
Emerging Forms of Representation: Recovering Autonomy, Making Proposals
The PT’s historically close ties to many of Brazil’s most active and visi- ble CSOs meant that its crisis came closer to them than a crisis in HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 21 another party would have (Flynn 2005). CSOs helped to form the PT as an instrument to challenge first military government and later, socioe- conomic inequality (Keck 1992). At subnational levels, the party had provided them with unusual opportunities for political participation (Baiocchi 2003; Nylen 2003). CSOs and the PT had joined in numerous campaigns, from those for direct elections of the president in the 1980s to antiprivatization efforts in the 1990s. CSOs welcomed and worked toward Lula’s election in 2002 (Hochstetler forthcoming). Four years later, the Coordination of Social Movements, a grouping of CSOs that had been historically close to the PT, deliberately withheld direct sup- port for the first round of Lula’s re-election campaign, but did take to the streets against the PSDB-PFL coalition that opposed it. In a morning of speeches on June 28, 2006, organizations like the CUT trade union, the MST landless movement, and the World March of Women insisted that only their mobilization could change economic policy, whatever the electoral outcome (personal observation; Barbosa 2006; Coelho 2006; Marcelino 2006). This argument, that CSOs alone would and could articulate an alter- native development model, had been evolving for some time. Before the PT reached the presidency, it was the highest-level aggregator of such claims, in close connection with CSOs (Nylen 2003; Cabral 2005). As Lula’s 2002 campaign progressed, with numerous reassurances that he would continue Cardoso’s neoliberal economic policies, however, a large group of CSOs formed the Brazilian Budget Forum (FBO) in August (before the October election) to create a voice for higher social spending. They also stressed the need for—and provided—greater public debate over budgetary issues, distributing 30,000 copies of a booklet describing concepts like the primary budget surplus in accessi- ble terms (FBO 2004). Numerous, partially overlapping networks of NGOs prepared alternative economic proposals: the Cry of the Excluded, the Coordination of Social Movements, the Brazilian Budget Forum, the Popular Assembly, and others. At the beginning of Lula’s administration, they debated these proposals in government-organized forums, such as the Economic and Social Development Council and the administration’s formal consultations on its “Multiyear Plan 2004–2007: A Brazil for Everyone.” Over several stages, however, the CSOs with- drew from these efforts to concentrate on extragovernmental initiatives, arguing that their efforts were wasted inside (Hochstetler forthcoming). The problem, from the CSOs’ point of view, was not the lack of an audience. The Lula administration was anxious to listen to them, and praised their efforts. Lula himself has met regularly with CSOs through- out his presidencies and has stressed their importance as voices for soci- etal interests and values (Abong 2003). The problem was that CSOs’ articulation and aggregation of interests did not make an impact on the 22 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 administration’s governing choices. Tens of thousands of organizations in 33 networks that make up the Inter-Redes umbrella group came to this conclusion when they formally disavowed the Multiyear Plan they had worked on (Inter-Redes 2004). This outcome shows the importance of political parties’ continuing dominance of the governing space. CSOs repeatedly tried during this time to catalyze their other poten- tial audience, the mass public, into action, in hopes of using broader public demands to push their economic vision. One place to measure their success in reaching this audience is in the annual Cry of the Excluded (Grito dos Excluídos), a large march held on September 7, Independence Day. Organizations close to the Catholic Church began the first Grito in 1995, and it spread across the Americas in 2000. A high point for these mobilizations was in 2002, when organizers counted 214,000 marchers and more than 10 million participants in a plebiscite on the Free Trade Agreement of the Americas. More than 98 percent of the voters rejected the treaty (Grito Dos Excluídos Continental 2002). In 2006, given the crises in Lula’s first term, organizers hoped for a massive mobilization of a million or more to draw attention to their demands for economic and political transformation. They set stringent rules prohibiting politicians from speaking. Estimates for actual partici- pants varied from organizers’ initial optimistic count of three hundred thousand (Grito Dos Excluídos Continental 2006) to the Estado de São Paulo’s (September 7, 2006) counts, which did not total ten thousand. Brazil’s poor, a major audience for whom these organizations claim to speak, formed Lula’s strongest voting bloc in his successful re-election effort (Hunter and Power 2007), rather than “voting with their feet” and swelling these protests. Indeed, Hunter argues that the PT leadership made numerous choices after 1994 that disappointed their CSO allies, because they calculated that the party’s historic radicalism could not deliver the majority electoral results they needed, especially from Brazil’s majority poor population (Hunter 2007, 458–59). Brazilian CSOs have struggled to muster their audiences on eco- nomic issues since well before the recent party crisis. The most signifi- cant change in their agenda with the crisis is the renewal of a significant political reform initiative. Because many organizations and personal activist histories originated during the struggle for democracy, they have a longstanding interest in political institutions and political reform. In the 1990s, much of this agenda focused on efforts to gain and consoli- date CSO and popular representation in governmental decisionmaking, such as the ubiquitous councils or participatory budgeting processes (Baiocchi 2003; Friedman and Hochstetler 2002). These efforts often counted the PT as collaborator or even initiator. CSOs’ experience with the PT in national office, especially after the cor- ruption crises, led them to rethink this focus. In Lula’s first three years HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 23 in office, almost two million people participated in his government’s national conferences on different issues (Reforma Política 2006, appen- dix 3); but as noted, CSOs were nearly always disappointed with the substantive outcomes. Hence, participation in government-established processes no longer seemed adequate for representing a full set of inter- ests and values. Initial reform efforts began in 2004 and broadened after the cor- ruption crisis. Numerous gatherings of organizations criticized represen- tative democracy and called for direct political mechanisms, albeit in fairly vague terms (Hochstetler forthcoming). Out of this generalized dissatisfaction, one group of CSOs took the initiative to work on more specific proposals. From August to November 2005, 60 participants rep- resenting 21 states discussed strategies to address the political crisis. As they themselves pointed out, their draft political reform proposals were considerably broader than the parties’ parallel reform process, which focused only on party and electoral reforms (Reforma política 2006, 25). The CSOs also proposed numerous reforms that would change the way representative democracy functioned—from term limits to the end of secret votes—but made additional proposals for direct democracy, par- ticipation, democratization of information and communication, and judi- cial transparency. Over the next year, extensive discussions at the Brazilian Social Forum and numerous other, geographically based forums brought many participants together to discuss and refine the proposals. They ignored the electoral calendar in these discussions, choosing to have their final meeting at the end of November 2006, when all the elections would be over. In March 2007, the CSOs presented their proposals to the newly elected congress and gained a set of parliamentary supporters. Since then, reform efforts increasingly have followed a traditional lobbying strategy, with parliamentary representatives raising agenda items in the congress for their CSO allies. In July 2007, the first of these, an effort to reform Brazil’s open-list electoral system, failed. Almost all PT repre- sentatives voted for the proposal, but members of their coalition parties helped to defeat this reform of partisan representation (Vigna 2007). There were no large supportive mobilizations. As these reform proposals suggest, Brazilian CSOs have no inten- tion of replacing representative democratic institutions, such as political parties and legislatures. CSOs want such institutions to work better, and they have many ideas about how to improve their functioning. At the same time, CSOs are adamant that political parties cannot monopolize the exercise of popular sovereignty. They want considerably more direct participation for individual Brazilian citizens. They also foresee an ongo- ing need for pressure from organized civil society regardless of which parties win elections. They have not been able to muster the support for 24 LATIN AMERICAN POLITICS AND SOCIETY 50: 2 significant outsider challenges—which is not surprising in a country where the party crisis was short and contained.
CONCLUSIONS
Political parties leave many gaps in representation. In the three coun- tries under study, these gaps are always present, but recent party crises sharpened citizens’ sense that their political parties do not adequately represent them. In all three countries, CSOs stepped in to fill some of the gaps, although they did so in different ways. In Bolivia, outsider CSOs have presented themselves as more authen- tic and direct representatives than parties, attempting to change the con- stitution to ensure a continuing role for themselves as political media- tors—or rejecting the political system altogether. In Argentina, some citizens formed outsider CSOs to sidestep partisan representation, reject- ing the political class in favor of directly representing themselves, while others focused on reforming political institutions. Over just a few years, however, the main party has reasserted its political dominance, including its traditional close relationship with many civil society actors. Other CSOs continue to work for reform, now often parallel to partisan representa- tives. In Brazil, CSOs never presented themselves as an organizational alternative to parties, but they worked hard as insiders to represent ideas and values that were not otherwise being represented. These included both substantive economic proposals and proposals to improve the polit- ical process, including the process of partisan representation. In its broad outlines, the crisis of representation hypothesis is sup- ported by these three cases. Bolivia, with the most severe party crisis, also sees the most extensive development of outsider strategies by CSOs, across the political spectrum. CSOs’ protests and demands increasingly dominate political outcomes. In Argentina and especially Brazil, lower levels of party crisis are associated with CSO strategies oriented more toward working within institutions and with at least some partisan rep- resentatives. Even in these cases, both citizens and governments have recognized that CSOs speak for important ideas, and have sometimes responded to their proposals. CSOs have given citizens another avenue for representation in the political system, well beyond that expected by pluralist theories. This representation expands when parties are in crisis and takes the form of alternative mechanisms through which citizens can influence political outcomes in stabler periods. The three cases also provide suggestions about the process through which the crisis of representation hypothesis works. In all three coun- tries, existing and new CSOs moved quickly to initiate outsider strate- gies in the wake of party crisis; CSOs were eager to expand the scope of their representational roles. The critical variable in determining CSOs’ HOCHSTETLER AND FRIEDMAN: CIVIL SOCIETY 25 representational “success” seems to be less the CSO initiatives than the popular and partisan response, particularly where party systems re- equilibrate relatively quickly. Brazilian CSOs were never able to garner strong support from the populace—their most important audience—for their initiatives, and such support also dropped off quickly in Argentina. Even many insider initiatives found little resonance. Without strong and mobilized support, their political interventions were received politely by (resurgent) partisan representatives, but to little substantive effect. The crisis of representation hypothesis, however, is far from adequate to explain developments in political representation in Latin America. It does not even contemplate the Latin American “third way” of populism, where neither parties nor CSOs are able to represent citizens. It also cannot account for the significant variations in CSO activities over time and space in Latin American democracies (Arce and Bellinger forthcoming), espe- cially those that are not in crisis. There are cycles of CSO activities that are not explained by moments of partisan or even regime crisis. All of these phenomena deserve further empirical and theoretical attention. Perhaps most seriously, the crisis of representation hypothesis suf- fers from the normative assumption that a substitutive expansion of CSO representation is necessarily a solution to the crisis of partisan repre- sentation. It could be, especially in presenting an alternative form of peaceful political organization when parties are deeply disappointing to citizens (Hochstetler forthcoming). But the Bolivian case shows some of the risks in greatly expanding mechanisms of representation that do not have clear rules for aggregation and resolution when there are deep national divisions. Since 2000, Bolivian political representation has often pitted the informal multitude against the formal assembly, and the informal multi- tude commonly won. Now that a multitude is often pitted against a mul- titude (“there must be indigenous autonomy, but not regional”; “no, there must be regional autonomy, but not indigenous”), the possibility of peaceful resolution of political conflict through collective decision- making by anyone looks ever more remote. Of course, each multitude has some partisan representation, and those actors appear equally unable to reach a peaceful resolution. In addition, parties’ historic inability to resolve Bolivia’s problems in inclusive ways set up this ever- deeper crisis. The final point is only that representation by CSOs does not appear to have resolved it.
NOTES
We would like to thank William C. Smith and three anonymous reviewers for many helpful comments on earlier versions of this article. Any remaining problems are, of course, our responsibility. We also thank Silvia Ostrovsky for research assistance. 26 LATIN AMERICAN POLITICS AND SOCIETY 50: 2
1. We are indebted to an anonymous reviewer for helping to clarify this point. 2. The Bolivian Constitution has provided that the congress will select the president from among the candidates if none has received a majority of the pop- ular vote. 3. We use the Bolivian term political force (fuerza política) to describe Podemos because it has the legal form of the “citizen grouping” described below. 4. Others argue that the Peronist vote is not stable and that the party system has been decentralized (Calvo and Escolar 2005). 5. Some observers have found that pre-existing political networks fostered these mobilizations (Auyero 2007). 6. As in Bolivia, the rise of direct protest as a strategy extended to the polit- ical right as well, most notably in the “law and order” mobilizations organized by the entrepreneur Juan Carlos Blumberg, whose son was kidnapped and murdered in March 2004.
REFERENCES
Abong. 2003. Lula recebe o novo conselho diretor da Abong. Informes Abong 246.
Bartolini, Stefano, and Peter Mair. 2001. Challenges to Contemporary Political Parties. In Political Parties and Democracy, ed. Larry Diamond and Richard Gunther. Baltimore: Johns Hopkins University Press. 327–43. Betancur, Ana Cecilia. 2004. Diez temas de reforma constitucional. Santa Cruz, Bolivia: Centro de Estudios Jurídicos e Investigación Social. Bielsa, Rafael A., Miguel Bonasso, Stella Calloni, et al. 2002. Qué son las Asam- bleas Populares. Buenos Aires: Ediciones Continente. Bolivia. Corte Nacional Electoral. 2005. Elecciones generales: votación para la elección de presidente, vicepresidente, senadores y diputados plurinomi- nales. Elecciones generales: votación para la elección de diputados uninom- inales.
DataFolha. 2005. Online dataset.
––––. Forthcoming. “We Could Make a Real Leftist out of Lula”: Organized Civil Society in Lula’s Brazil. In Democratic Brazil Revisited, ed. Peter Kingstone and Timothy J. Power. Pittsburgh: University of Pittsburgh Press. Hunter, Wendy. 2007. The Normalization of an Anomaly: The Workers’ Party in Brazil. World Politics 59, 3 (April): 440–75. Hunter, Wendy, and Timothy J. Power. 2007. Rewarding Lula: Executive Power, Social Policy, and the Brazilian Elections of 2006. Latin American Politics and Society 49, 1 (Spring): 1–30. Inter-Redes. 2004. Comunicado Inter-redes e ABONG.
Crisis of Democratic Representation in the Andes, ed. Mainwaring, Bejarano, and Pizarro Leongómez. Stanford: Stanford University Press. 1–44. Manin, Bernard. 1997. Principles of Representative Government. Cambridge: Cambridge University Press. Marcelino, María Fernanda. 2006. Member, Sempreviva Feminist Organization (SOF) and World March of Women. Author interview. São Paulo, June 29. Mayorga, René Antonio. 2005. Bolivia’s Democracy at the Crossroads. In The Third Wave of Democratization in Latin America: Advances and Setbacks, ed. Frances Hagopian and Scott Mainwaring. Cambridge: Cambridge Uni- versity Press. 149–78. Nylen, William R. 2003. Participatory Democracy versus Elitist Democracy: Lessons from Brazil. New York: Palgrave Macmillan. Ojeda, Julieta. 2007. Coordinator, Mujeres Creando. Author interview. La Paz, June 13. Pacto de Unidad. 2004. Una asamblea constituyente de todos, con todos y para todos. Mimeograph. Santa Cruz, Bolivia: Pacto de Unidad. Pearce, Jenny. 2004. Debate: Collective Action or Public Participation? Comple- mentary or Contradictory Democratisation Strategies in Latin America? Bul- letin of Latin American Research 23, 4 (October): 483–504. Peruzzotti, Enrique. 2002. Towards a New Politics: Citizenship and Rights in Contemporary Argentina. Citizenship Studies 6, 1 (March): 77–93. ––––. 2006. Representation, Accountability, and Civil Society. In NGO Account- ability: Politics, Principles, and Innovation, ed. Lisa Jordan and Peter Van Tujil. London: Earthscan. 43–60. Pitkin, Hanna Fenichel. 1967. The Concept of Representation. Berkeley: Univer- sity of California Press. Podemos. n.d. Proyecto de constitución política del estado: constituyente para la gente, propuesta para el debate. Mimeograph. No location. Powell, G. Bingham. 2004. Political Representation in Comparative Politics. Annual Review of Political Science 7 (May): 273–96. Przeworski, Adam, Susan C. Stokes, and Bernard Manin, eds. 1999. Democracy, Accountability, and Representation. Cambridge: Cambridge University Press. Ramiro Ballivián, Salvador. 2006. Análisis de la elección de la asamblea consti- tuyente y del referendum sobre autonomía departamental. In Opiniones y análisis: estudio de los resultados de la asamblea constituyente y el referen- dum sobre autonomías departamentales, ed. Ivonne Fernández Weisser. La Paz: FUNDEMOS. Reforma política: construindo a plataforma dos movimentos sociais para a reforma do sistema político no Brasil: reflexões para o debate. 2006.
Roberts, Kenneth M. 2002. Party-Society Linkages and the Transformation of Political Representation in Latin America. Canadian Journal of Latin Amer- ican and Caribbean Studies 27, 53 (January): 9–34. Saba, Roberto. 2006. Executive Director, Association of Civil Rights (ADC). Author interview. Buenos Aires, July 12. Samuels, David. 2006. Sources of Mass Partisanship in Brazil. Latin American Politics and Society 48, 2 (Summer): 1–27. Sánchez, Omar. 2005. Argentina’s Landmark 2003 Presidential Election: Renewal and Continuity. Bulletin of Latin American Research 24, 4 (October): 454–75. Sanchís, Norma. 2006. Regional Coordinator, International Network of Gender and Commerce. Author interview. Buenos Aires, July 12. Saward, Michael, ed. 2001. Democratic Innovation: Deliberation, Representation and Association. New York: Routledge. Schvartzman, Elsa. 2007. Member, Forum for Reproductive Rights and National Campaign for the Right to Legal, Safe, and Free Abortion. Author interview. Buenos Aires, August 6. Sitrin, Marina, ed. 2005. Horizontalidad: voces de poder popular. Buenos Aires: Cooperativa Chilavert Artes Gráficas. Smith, William C., and Roberto Patricio Korzeniewicz. 2007. Insiders, Outsiders, and the Transnational Politics of Civil Society. In Governing the Americas: Assessing Multilateral Institutions, ed. Gordon Mace, Jean-Philippe Thérien, and Paul Haslam. Boulder: Lynne Rienner. 151–72. Smulovitz, Catalina, and Enrique Peruzzotti. 2000. Societal Accountability in Latin America. Journal of Democracy 11, 4 (October): 147–58. Svampa, Maristella. 2005. La sociedad excluyente: la Argentina bajo el signo del neoliberalismo. Buenos Aires: Taurus. Svampa, Maristella, and Sebastián Pereyra. 2003. Entre la ruta y el barrio: la expe- riencia de las organizaciones piqueteras. Buenos Aires: Editorial Biblos. ––––. 2005. La experiencia piquetera: dimensiones y desafíos de las organiza- ciones de desocupados en Argentina. Medio Ambiente y Urbanización 61, 1 (February): 97–112. Taylor-Robinson, Michelle M. 2001. Old Parties and New Democracies: Do They Bring out the Best in Each Other? Party Politics 7, 5 (September): 581–604. Torre, Juan Carlos. 2003. Los heurfanos de la política de partidos. Sobre los alcances y la naturaleza de la crisis de representación partidaria. Desarrollo Económico 42, 168 (January–March): 647–65. Turner, Frederick, and Marita Carballo. 2005. Argentina: Economic Disaster and the Rejection of the Political Class. Comparative Sociology 4, 1–2 (March): 175–206. Valente, Marcela. 2006. Argentina: Human Rights Trials Trigger Wave of Threats. Inter Press Service News Agency, October 4. Van Cott, Donna Lee. 2005. From Movements to Parties in Latin America: The Evolution of Ethnic Politics. Cambridge: Cambridge University Press. Verdú, María del Carmen. 2006. Spokesperson and lawyer, Coordination Against Police and Institutional Repression (CORREPI). Author interview. Buenos Aires, July 10. Vezzulla, Juan Martín. 2006. Legal Adviser, Environment and Natural Resources Foundation (FARN). Author interview. Buenos Aires, July 12. 32 LATIN AMERICAN POLITICS AND SOCIETY 50: 2
Vigna, Edélcio. 2007. O futuro da reforma política.